The Stateless Polis: A Reply to Critics

The Stateless Polis: A Reply to Critics The Stateless Polis: A Reply to Critics
Author: Berent, Moshe
Journal: Social Evolution & History. Volume 5, Number 1 / March 2006


In a series of articles based on my Cambridge doctoral thesis (Be-rent 1994) I have argued that, contrary to what has been traditionally assumed, the Greek polis was not a State but rather what the anthropologists call ‘a stateless society’. The latter is characterized by the absence of ‘government’, that is an agency which has separated itself out from the rest of social life and which monopolizes the use of violence. In stateless societies the ability to use force is more or less evenly distributed among the armed or potentially armed members of the community (Fortes and Evans-Pritchard 1940). In this series of articles I have also tried to show how the analysis of the Greek polis as a stateless society leads to a better understanding of both the Greek polis and Greek political theorists (Berent 1996: 36–59; 1998: 331–362; 2000a: 257–289; 2000b: 228–251; 2000c: 2–34; 2000d: 225–241; 2005: 364–387).

The purpose of the present paper is to respond to critical essays by Grinin and van der Vliet which have been published recently in Social Evolution and History and which have rejected the notion of the stateless polis (Grinin 2004: 93–147; van der Vliet 2005: 120–150). I have already related to some of the criticisms in my rejoinder to a critical essay by M. H. Hansen (Hansen 2002: 17–47), and some of the arguments employed here were first elaborated in this rejoinder (Berent 2004b: 1–2).


One of the major criticisms which has been leveled against my thesis concerns my choice of the definition of the State. I was accused by Hansen that my definition of the state is too wide and unspecified (Hansen 2002: 20), while both Vliet and Grinin accuse me of using a too narrow concept of the State. According to Vliet I am ‘using the restricted, hobbesian definition of the state in the study of the Greek polis’ (van der Vliet 2005: 133). According to Grinin my model of the early State singles out only the bureaucratic or monarchic early State while it does not give enough consideration to ‘the specific character of democratic states’(Grinin 2004: 100–101). As for the emphasis which I put on what I call Weber's definition of the State, while Grinin and Vliet agree that the execution and maintenance of the laws were private, they do not consider it as an indication of the statelessness of the polis. Thus it is argued by both Grinin (p. 119) and Vliet (p. 134) that by this the polis resembles the early modern State, where the role of self-help in the maintenance and the execution of the laws was crucial. This argument was advanced also by Hansen and I have related to it in some length in my rejoinder.

Yet, I believe that this criticism is based on a misunderstanding of the way that I use the Hobbesian or Weberian concept of the State. It should be emphasized that for establishing the statelessness of the Greek polis we do not need a clear or an agreed definition of the State, but rather what we need is a definition of a stateless society. Thus I define a stateless community as a community without coercive apparatuses, or as a community in which the ability to use force is more or less evenly distributed among armed or potentially armed members of the community. While in my definition of the stateless society I am making use of the Hobbesian or Weberian definition of the State, this does not mean that I am necessarily committed to it or to any other definition of the State. Thus I do not define a stateless society as a society which do not posses a monopoly of coercive apparatuses, as Grinin and Vliet seem to suggest, but rather as a society which does not have coercive apparatuses at all. Grinin and Vliet are certainly right to point out that while the Agrarian and the early (modern) state did not possess a monopoly of violence we would still incline to call them ‘states’ and that consequently the absence of a monopoly of violence does not necessarily imply the absence of a State (see also Gellner 1983: 3–5). Yet, unlike what they seem to suggest, those early modern States were still different from the Greek polis in the sense that they were relatively centralized and they possessed coercive apparatuses. Similarly it would be a mistake to equate the Greek polis to the late mediaeval Italian city-states. While it could be doubted by some whether the latter could be properly termed ‘states’, they were certainly not stateless. Unlike the Greek polis, the Italian city-state of the late Middle Ages had State-coercive apparatus in the form of militias which were available for internal policing duties (Berent 1996: 43–45). In other words, early modern States had a political centre, though it might not rule effectively as Weber's definition demands, and anyone who wished to rule these States had to take over this political centre. The polis, like any other stateless community, was decentralized and did not have a political centre. Thus there were no institutions that one needed to take over if he wanted to rule, no existing power structure and institutions of government. In other words, the polis was ungovernable, unless the governor (let us say the tyrant) created his own administration (and see also Berent 2004b: 117–119). Vliet seems to think that the polis did have a political centre arguing that the polis, though a State, was different from the early state as it was ‘a democratic state’ and consequently ‘the political system of the polis has a flat structure. Its centre is not above, but in the midst of its citizens’ (p. 141). I will return to this point later on.

Thus I do not believe that in order to establish the statelessness of the Greek polis we need a thorough discussion of definitions or characterizations of States, ancient and modern. Vliet is wrong when he maintains that the main foundation of my discussion of the statehood of the polis is Aristotle's analysis (p. 133). This is certainly not how I see it. My discussion of the statelessness of the Greek polis starts with the assumption which I have all the reasons to believe that it is unobjectionable, namely the absence of coercive apparatuses in the Greek polis. And, yes, I believe that this absence is reflected by Greek political theorists in General and Aristotle in particular. Of course, by saying that the polis was stateless I also say that it was not a State. Yet, all the theorists of the State listed by Vliet in the opening of his paper (pp. 121–124), while they could have different opinions about the question what is the State, would probably agree that a society which does not possess coercive apparatuses is not a State. And I believe that both, Grinin and Vliet would also ascribe to it, for they both, wrongly from my point of view, seem to point at the existence of some degree of centralization and coercion in the Greek polis as a proof of its statehood. Thus when Grinin says that ‘a developed apparatus of coercion is not strictly obligatory for an early state’ (p. 120, emphasis added) I take it that he means that some sort of an underdeveloped apparatus of coercion is obligatory and that the polis had such an apparatus and that consequently he would agree that a society without coercive apparatuses is not a State. Indeed Grinin seems to advance the notion that the polis was a State with a weak executive branch. Similarly, Vliet says that ‘The presence of certain institutions, which are permanent and through which legitimate force can be used, proves in my opinion the presence of the state’ (p. 123) and that ‘(d)ecisive, in my opinion, is the Weberian principle of the legitimacy of the monopoly of force, which also can be found in the midst or complementarity of the institutions’ (p. 124). Thus, when I read the arguments employed by both Vliet and Grinin I tend to think that they would also agree that a society without coercive apparatuses is not a State, though they consider the polis to posses such coercive apparatuses.

Vliet and Grinin's treatment of the issue of the existence of a monopoly of violence in the Greek polis point also at another misunderstanding of the way which I use facts about the polis. ‘Berent's conclusion that there was no state if the citizens could manage slave exploitation themselves is illegitimate’ (Grinin, p. 108). Grinin is absolutely right when he says that one cannot infer the absence of the state from the fact that the slaves were controlled by self-help. Yet he is wrongly accusing me of saying so. As David Hume had taught us theories could not be proved by or deduced from facts. On the contrary, the fact that slaves were managed by self-help is to use Karl Popper terminology a ‘falsifying theory’ or a ‘corroboration’ of the statelessness of the Greek polis. Had Greek slavery been regulated by (collective or public) coercive apparatuses this would have been a refutation of the theory of its statelessness. The fact that the regulation of slaves had been by private means only corroborates the statelessness of the Greek polis, it doesn't prove it. Similarly the fact that court's judgments were executed by private means is not a proof but rather a corroboration of the statelessness of the polis.

The Polis and Coercion

So my first task would be to look at the arguments employed by Grinin and Vliet which, according to them, show that coercive apparatuses existed in the Greek polis. Grinin argues that in the polis ‘The repressive apparatus was also available … the total number of police sub-units was initially 300 and later 1200’ (p. 127). Similarly Vliet says that ‘(i)t is certain, however, that the Athenians had a police force of public slaves, with bows armed Scythians’ (p. 128). Here, I believe, that I am treading on a rather safe ground. That the Greek polis had never developed a police system seems to me to be the accepted view. The duties of the three hundred Scythian slaves mentioned by both Grinin and Vliet was to keep order in the Assembly and the courts; they were also at the disposition of several boards of magistrates. Yet, there is no evidence that they have been used as a policing force. The only thing which resembles State coercive apparatus in the Athenian polis were ‘The Eleven’ (hoi hendeka) which were responsible for the prison and for summarily executions. Yet, given the number of inhabitants in Attica (over 200,000) this element is indeed redundant (Berent 1996: 40–42).

Another coercive apparatus according to Grinin was the army. Thus he argues that ‘in Athens and other poleis at the beginning of the second half of the 5th century B. C. the contingent of mercenaries was growing up and later it become the dominating one’ (p. 113). Indeed, to a certain extent the use of mercenaries contradicts the idea of the polis as a community of non-professional warriors, but this does not necessarily contradict the notion of its statelessness. For mercenaries could be used either for ruling or for waging war against the outside world. To the extent that they were used for war against the outside world then their function was mainly to supplement for the polis's shortage of warriors, but this would not have made them into a ruling apparatus (Berent 2004b: 115–117). On the other hand, sometimes mercenaries were employed by tyrants for the purpose of ruling, yet tyrannies were indeed attempts to centralize power and, as Finley notes, the very negation of the idea of the polis (see also Berent 2004b: 116–117). Grinin does not seem to distinguish between the use of (collective) force towards the outside world, on the one hand, and the use of (collective) force for the purpose of ruling, on the other. It is only the latter which is absent from stateless society while the former in fact characterizes it. Thus while Grinin is right when he says that ‘wars were of great importance in the formation early states (p. 108)’, protection from the outside world (or predation) could acquire the form of a stateless society. Indeed, stateless communities tend to be communities of warriors or communities with ‘high military participation ratio’ and wars had played an important role in the maintenance of the economy and the social cohesion of stateless communities in general and the Greek polis in particular (Berent 2000a).

Bureaucracy and Differentiation

Complementary to the absence of coercive apparatuses in the Greek polis was the absence of differentiation of a ruling class or a bureaucracy. As Gellner notes:

... the state is the specialization and concentration of order maintenance. The ‘state’ is that institution or set of institutions specifically concerned with the enforcement of order (whatever else they may be concerned with). The state exists where specialized order-enforcing agencies, such as police forces and courts, have separated out from the rest of social life. They are the state (Gellner1983: 4).

As Gellner notes, in agrarian States the differentiation of a bureaucracy acquires the form of class-differentiation.

On the existence of a bureaucracy the position of both Grinin and Vliet seem to somewhat unclear. On the one hand they both maintain that the number of bureaucrats in the Greek polis was small (Grinin, p. 115, Vliet, pp. 127–128), yet they both seem to ascribe to the idea that the polis-institutions could be viewed as a state-administration. Grinin says that ‘the state apparatus was available … in the polis … though it was of a specific type’ (p. 115). On p. 125 we encounter the following

Though the apparatus of management and coercion in Athens and Rome was not so powerful as in bureaucratic countries, it was quite still numerous, especially in Athens. In the 4th century B. C. there lived only 200,000 people. At the same time many hundreds of citizens were directly involved in the administration (being elected or chosen by lot).

If one adds the numbers of the people which staffed the polis-institutions on Grinin's list, the number of people participating in the administration is indeed several thousands. Given the ratio between the number of the latter and the total number of people who lived in Attica we face a rather strange conclusion. To the extent that the polis was traditionally described as a State, it was agreed that it was, as Finley put it, a ‘non-bureaucratic State’ (Finley 1985a: 151 – emphasis added). Now according to Grinin, not only that the polis was a State, but if we allow for ‘state apparatus … of a specific type’, then it was the most (specific type) bureaucratic state ever to exist even when compared to the early ‘bureaucratic States’ which Grinin mentions in the quotation above as more bureaucratic then the Greek polis.

In other words, the problem with Grinin argument is that it does not distinguish between participation and bureaucracy. For, if the popular courts are included in the ‘administration’, why not include the assembly itself and then we would have a complete identity between the ‘administration’ and the community.

The question which arises here is how one distinguishes between bureaucracy or administration, on the one hand, and participation, on the other. I believe that the traditional answer was ‘differentiation’. Here Grinin brings the quotation by Gellner (which I have myself used frequently):

... the state is the specialization and concentration of order maintenance. The 'state' is that institution or set of institutions specifically concerned with the enforcement of order (whatever else they may be concerned with). The state exists where specialized order-enforcing agencies, such as police forces and courts, have separated out from the rest of social life. They are the state (Gellner 1983: 4).

While first allowing that ‘Gellner's idea make some sense’ (p. 132), Grinin's analysis makes it in fact senseless. Referring to this quotation Grinin says that ‘in this respect the ancient societies give us a good example that we are dealing with real states. While police was not a very important body, the court achieved a high degree of development and significance’ and that in Athens ‘it is possible to consider courts as apparatus of coercion, as they gave sanction to apply force though they quite often left it to the interested part itself to execute the judgment’ (p. 114). These statements are a complete misunderstanding of what Gellner says for two reasons. First, the police is indeed important and Gellner would have certainly not considered a society with courts which do not have coercive apparatuses at their disposal as a state. Both states and stateless societies sanction the use of force, yet in the latter force, though sanctioned, is predominantly private. Second, Grinin completely ignores the issues of specialization and differentiation, that is ‘the separation from the rest of social life’. Thus seeing the popular courts in Athens as differentiated bodies is highly problematic. According to Gellner in agrarian Societies differentiation acquires the form of professionalism, on the one hand, and class differentiation, on the other. Both are absent in the Greek case as the courts were popular and staffed by amateurs. Here Grinin claims that courts in the (modern) State are sometimes staffed by amateurs as well (as in jury trials) (p. 117). Yet in the modern case the differentiation of the courts from the society is done also through the existence of professional judges and intermediately professional bureaucracy and police which runs the courts and carries out the courts’. This was also absent in the Greek case. Thus polis-institutions, in as much as they had been an elaborate and complex system were still not differentiated or separated from the rest of social life. Grinin ignores the issue of differentiation to the extent that it enables him to speak about ‘separation of powers’ in the Greek polis

… though the evolutionary theory triumphed, that is the theory which put an emphasizes that state administration is formed of professional officials and among the branches of power the executive one becomes the main, it does not mean that there are not other possibilities. The polis is one of them. As it was a democratic state, it is natural that the legislative and judicial powers were more developed there. So the civil executive power could be weak. The executive military power was much stronger.

Similar arguments advocating the separation of powers in the Greek polis had been advanced by M. H. Hansen. Yet, separation of powers is possible only when there is a distinct or differentiated ‘power’ which could be divided. It was J. Ober who has pointed out that ‘Hansen's strong arguments for the prevalence of amateurism in Athenian Government tend to weaken his thesis concerning the separation of powers’ (Ober 1989: 327).

Grinin takes the issue of differentiation to another direction when he points out that ‘professional politician were available’ seeing it as a manifestation of the polis statehood (p. 115). Indeed, in Athens it is possible to distinguish between ‘government’ in the sense of political institutions and officials, on the one hand, and ‘government’ in the sense of people who formulated policy, on the other. While the political institutions and offices were staffed by amateurs, thus exhibiting no division of labor between rulers and the ruled, one can speak of a certain kind of a division of labor considering the ‘professional politicians’ in Athens, that is the demagogues and those who proposed and spoke in the assembly. Yet, while as Gellner notes, a division of labor is a necessary condition for the existence of the State, it is certainly not a sufficient one. Thus, in the sense that the professional politicians in Athens could be called ‘a government’, this was certainly a non-State government. The Athenian leader did not have any formal position or State coercive apparatus at his disposal. He was simply a charismatic individual, a demagogue, who could persuade the people in the Assembly to accept his policies, but still risked losing his influence (and his life!), and having his policies rejected at any moment (Finley 1985: 24). The existence of both types of ‘government’, formal and informal, characterizes stateless communities. While in State societies leaders must be part of the State-hierarchy, in a stateless community alongside with formal positions (such as chiefs) there are powerful individuals who have no formal position – such is ‘The Big Man’ – ‘who does not come into an existing position of leadership over a certain group but personally acquires dominance over certain other fellows’ (Sahlins 1968: 22).

Vliet's notion of bureaucracy and differentiation in the Greek polis is not very much different from Grinin's though he seems to be more careful. While he quotes Vincent with approval that ‘Historically and anthropologically it is clear that both the concept and practice of government existed before the State. Government can and does exist without the State’ (p. 122), he refrains from making one step further to suggest that the Greek polis offers such example. Thus he speaks of ‘the complexity of administrative apparatus’ in Athens (p. 129). Like Grinin he points at the large scope of participation in the government, yet he understands that in order that this government would be indeed a State-government it ought to be supplemented by administration or bureaucracy, thus he maintains that ‘(i)t is impossible to imagine that this entire apparatus of government could have functioned without an administration’. Yet, he admits that ‘We do not know whether they or the other office-holders did have at their disposal a vast staff of administrative assistants, a kind of bureaucracy, or of how many people these consisted and of whom’ (p. 128). Indeed, both Grinin and Vliet seems to ascribe to what I call ‘the argument from complexity’, that is the idea that because polis institutions were complex, it must have been a State, and I will return to this argument later on.

The polis as ‘a democratic State’

In as much as both Grinin and Vliet consider the polis to be a State, they seem both admit that the traditional theories of the early State are not easily reconciled with the structure of the polis. Both seem to ascribe it to the fact that theories of the early State had focused on the bureaucratic authoritarian States while they seem to neglect the early ‘democratic states’, of which the polis is but one example. While this criticism might be true for the anthropological research of the early State, it is certainly not true in as much as the Classics are concerned. Here, the peculiar ‘democratic’ character of the Greek polis when compared to both ancient and modern States was not entirely ignored. What that had troubled the traditionalists was not the absence of coercive apparatuses, which they ignored, but the undifferentiated or ‘democratic’ nature of the Greek political institution. Assuming that the polis was a State, a set of propositions had to be devised, allowing for the Greek polis to be a State, on the one hand, yet still of a ‘specific kind’, on the other. Thus what I have termed as ‘The Traditional Theory of the Greek Polis’ had tried to overcome the issue of differentiation by suggesting that while the (modern) State is characterized by the separation or the distinction between State and Society, in the Greek polis there was a unity of State and Society (Barker 1960: 12). Another formulation, or corollary, of this basic tenet was the idea that while in the (modern) State there is a distinction between the state and the citizens, or between rulers and the ruled, in the Greek polis there was an identity between the State and citizens and no distinction between rulers and the ruled. Thus Ehrenberg says that ‘The Polis was the state of the politai, the citizens ... Andres gar polis (Thuc. Vii. 77. 7): it is the men who are the Polis. There were no subjects’ (Ehrenberg 1969: 88). Indeed, both Vliet and Grinin seem to ascribe to the notion of the ‘citizens-state’. Thus Vliet maintains:

In the citizens-states of classical Antiquity the state resides in its citizenship, and is expressed in its ideology of citizenship, the paticipation of the citizens in office-holding and decision-making on various level ... The citizenship constitutes the state, as well as participation in the state defines the citizen (p. 143).

Yet the notion of the Greek ‘citizens state’ is problematic from theoretical point of view and not so easily reconciled with the theory of the State. As A. P. d'Entrèves observes that ‘The word “State” is generally associated with the notion of a force outside the individual will, superior to it, but able not only to issue commands but to enforce them’ (d'Entrèves 1967: 1). Quentin Skinner refuses to see sixteenth- and seventeenth-century republican theories as theories of the State because they ‘make no ... distinction between the powers of the State and those of its citizens’ and this ‘is far from being the concept we have inherited from the more conservative mainstream of early modern political thought’ (Skinner 1989: 112–113). Indeed, it is not clear how Vliet reconciles the notion of the Greek ‘citizens state’ with his opening theoretical discussion where he quotes Vincent with approval that in states: ‘[the] public power is formally distinct from both ruler and ruled’ (p. 123). It was M. I. Finley who attacked this traditional theory as ‘vague’ and ‘metaphysical’ when he said

the notion that a state can be characterized – one might almost say defined – by the sovereignty of the citizen with full rights is only a short step from the nonsense of ‘das römische Volk ist der römische Staat’ (the Roman people is the Roman state) (Finley 1983: 8).

This observation had not led Finley to abandon the notion of the ‘Greek State’, but rather to impose differentiation and distinction between rulers and ruled on the Greek polis (see Berent 2004b: 135). Another possibility to solve the problem is to recognize that the polis was a stateless society as the notion of the latter is easily reconciled with that of the identity between the polis and its citizens. Nevertheless, Finley's attack upon the traditional theory was very much in place. Its ‘metaphysical’ somewhat obscure character is demonstrated by Vliet when he says that as opposed to the early authoritarian States ‘the political system of the polis has a flat structure. Its centre is not above, but in the midst of its citizens’ (p. 141). Thus it is not very clear what it means for a political center to reside ‘in the midst of the citizens’. As Finley, qouting Harlod Laski, noted

... in the study of politics, there is no meaningful distinction between state and government. Political metaphysicians notwithstanding, the citizens (or subjects) in any regime equate the two (Finley 1983: 8).

In other words, States do have tangible centers, diffused or decentralized as it may be in democratic states, it is still distinct from the citizens. Similarly, it is not clear what Vliet means when he says that the typical political structure of the classical citizens-states ‘hides much of the apparatus by which ... power was and could be administered’ (p. 143). In as much as (modern) democracies apply different forms of power, their coercive physical form of power is apparent. In another place Vliet says about the fact that also in public matters prosecution was initaited by priavte citizens (ho boulomenos) that

... considering this as a testimony to the absence of the state, is to deny the nature of Greek citizenship and the peculiar nature of the polis as a citizen-state. The individual acted as a citizen, that is he represented the common interest. ... it was not a separate organ of the state, but the individual citizens together who were responsible for the maintenance of the laws. That does not imply that that was less effective – rather the contrary, perhaps (p. 135).

It is indeed very difficult to find the State in all these quotations. Why call it a ‘peculiar nature of the polis as a citizen-state‘ when this feature – the fact the public interest is not represented by a special differentiated agency, but rather by ordinary members of the communities – is characteristic of stateless societies? Indeed, the notion of ‘democratic’ citizens-state employed by both Grinin and Vliet does not seem to be very helpful for the understanding of the Greek polis simply because what they single out as peculiar to the democratic early states when compared to early authoritarian states does not characterize (modern) democracies. Thus Vliet points out at ‘a remarkable paradox’:

On the one side it appears that how more democratic a state is, how weaker it is to enforce its power, because its rather strong dependency on the consent and cooperation of its citizens, who do not want to be forced, but who wish to be convinced. Plato complained that the Athenian democracy did not execute the penalties its courts imposed ... (p. 142)

Now, can we really say that the (modern) democracy is characterized by this paradox, that is that the more democratic it is, the less it is able to enforce its court's decisions? This is not a ‘paradox of democracy’ but rather a paradox which is associated with the notion of the ‘Greek State’. If you abandon the notion latter, the so-called ‘paradox’ disappears. Finley, who had noticed the absence of coercive apparatuses in the Greek polis, understood that this paradox is uniquely a Greek one:

Neither police action against individual miscreants nor crisis measures against large scale ‘subversion’ tells us how a Greek city-state or Rome was normally able to enforce governmental decisions through the whole gamut from foreign policy to taxation and civil law, when they evidently lacked the means with which, in Laski's vigorous language, ‘to coerce the opponents of the government, to break their wills, to compel them to submission’ (Finley 1983: 24).

Finley's attempt to struggle with this paradox was unsuccessful in my opinion, since he had refused to abandon the notion of the Greek State (Berent 2004b: 129–131). His solution was very similar to the one which is employed here by Vliet, that is the resort to legitimacy and consent (pp. 142–143). Yet, unlike Vliet, while resorting to legitimacy, Finley acknowledges that it has formed the basis ‘of many states in the past and present, though few (and perhaps none) with so little coercive power readily to hand’ (Finley 1983: 24 (emphasis added). Thus the polis is still unique and the paradox unsolved.

Grinin, following M. H. Hansen, seems to employ the notion of the citizens-state in a somewhat different manner. Like Finley, Hansen has attacked the notion of the identity between the State and the Society which was traditionally attributed to the Greek polis. According to Hansen

The polis was a society of citizens. It was a male society from which women were excluded; all foreigners were also excluded, and metics and slaves, though domiciled in the polis, were not members of it, a fact of which they were reminded every day of their lives, when the citizens went off on their own to deal with affairs of state in the Assembly or the Council or the courts. ... Yet every day, when the meetings to deal with affairs of state were over, citizen, metic and slave went off to work side-by-side as artisans, traders or farmers: in the economic sphere the stranger was a part of the society, though in the political sphere he was not (Hansen 1991: 62; 1998: 89–91).

Thus, according to Hansen the distinction between State and Society meant that ‘the polis, which was the political community of the citizens, and the society as a whole, in which all groups participated, were clearly distinguished’ (Hansen 1991: 64). This enables Hansen, pace the traditionalists, to retain both the notion of the citizens-state, on the one hand, and that of the separation between State and Society, on the other. Hansen's argument is repeated here by Grinin (pp. 112–113) and I have related to it in some length in my rejoinder. While both Hansen and Grinin are certainly right when they say that the polis as a political community or as a public sphere was distinguished from the whole society, or from the private sphere, they are wrong to imply that it was distinguished as a State. It seems that Hansen's interpretation would match also tribal societies which are clearly stateless. Tribal society is predominantly a male society. Thus within the territory which the tribe lives there are also ‘disenfranchised’ individuals such as women, children, strangers and also slaves. This would not make the distinction of the tribe as a male society from the rest of the population a differentiation of a State and it would not make the males ‘rulers’ over the rest of the population within this territory (Berent 2004b: 131–134). In short, what would have made the distinction of the citizens from the rest of the population a distinction of a State is either their being a professional ruling class, which they were not, or the existence of an intermediate bureaucracy between the various citizens assemblies and the whole population, which was absent.

The argument from complexity

One of the arguments which is commonly advanced by the proponents of the statehood of the Greek polis could be termed as ‘the argument from complexity’. According to this argument the complexity of the polis' society and institutions is an indication for the existence of a State. Hansen seems to advance a similar view when he says that in the Greek polis

… diplomacy and inter-poleis relations were formalized and institutionalized to an extent which makes it inappropriate to think of the poleis as stateless societies of the African, Pacific, or North American type … (Hansen 2002: 40–41).

This argument could be extended further to other areas: economy, law, the existence of cities, written language and many more. The argument from complexity seems to have two variants: The evolutionist and the functionalist. The evolutionist variant, employed here by Grinin, suggests that since the polis-institution offered ‘new’ or different modes of government then those of what are considered as ‘pre-state’ societies (and since the next evolutionary stage is statehood) the polis must have been a State (pp. 116–117). The functionalist variant of the argument from complexity says that complex societies could not function without a State. This approach is implied by Vliet when he says that ‘(it) is impossible to imagine that this entire apparatus of government could have functioned without an administration’ (p. 128) or that

… in the case of classical Greece ... we know that an offi-cialdom existed, but how it worked and was organised, how it was recruited, and how it was educated, escapes our know-ledge. But the argument e silentio should not be an argument against the statehood of classical political systems (p. 141).

Nevertheless, while it is true that the polis was a much more complex society in almost every aspect when compared to ‘primitive’ or tribal stateless communities, as far as its degree of centralization is concerned, the polis would rank fairly close to some ‘primitive’ tribal acephalous societies. In fact, some ‘primitive’ tribal societies, such as chiefdoms, show a greater degree of centralization than the polis (Sahlins 1968: 20). Indeed, I believe that Grinin is right when he says that ‘it is absolutely necessary to give up the unilinear approach to evolution in general and of the state in particular’ (p. 98). But this would mean also to abandon the argument from complexity. There is no reason to assume that since the polis had ‘new forms of government’, different from those which have been traditionally attributed to pre-state societies, then the polis must have been a State. One could imagine a development or ‘evolution’ of political institutions within a stateless framework. It is also possible for a society to be complex in some of its aspects and simple in others.

Indeed, the notion of the statelessness of the polis poses a problem to the traditional theory of the State which had suggested that complex societies must be instituted in State forms. The latter considered the stateless community to be simple or ‘primitive’, on the one hand, and the State as a necessary condition for civilized life, on the other (Sahlins 1968: 6–7; Khazanov 1978: 89–90; Crone 1986: 49–50). Behind this view was the idea that a civilized society which is both massive and divided within itself into classes, ethnic groups, specialized occupations and so on cannot stand without special means of control and integration. Or as Marshall Sahlins puts it ‘[t]he cultural richness that we call civilization has to be instituted in state form’ (Sahlins ibid.). Classical Marxism is a variant (or perhaps we should say the archetype) of this notion of the State. Both from an evolutionary and functional point of view it has suggested that wherever classes or private property could be discerned there must be a State. This argument seems to be used here by Grinin where he relates the existence of classes and exploitation in the Greek polis to its statehood saying that Athens ‘used the State for creating and keeping social and political inequality, economic exploitation and privileges of one Group over the other, and did it not worse but even better then other states’ (pp. 110–111).

Thus the ancient Greek world defies both modern political science and social anthropology by being both stateless and civilized: Greek civilization evolved in stateless conditions and was sustained by non-state mechanisms. Yet, if the anthropological view which identifies civilization with the State is not to be entirely rejected, then one must assume that the abilities of stateless communities to sustain an heterogeneous society are very limited and this could perhaps explain why the polis was, to use Runciman's phrases, ‘doomed to extinction’, or an ‘evolutionary dead-end’ (Runciman 1990: 348) and had eventually to make place for the large empires that followed (see Berent 2000a: 285–289).

One of the variants of ‘the argument from complexity’ could be termed as ‘the argument from effectiveness’. This argument suggests that since the polis was a stable peaceful political community, or since in the polis there was an effective rule of the law, then it must have been a State. Thus Grinin says that

It is not only the monopoly of application of coercion that is typical of an early state but the concentration of lawful application of force. It could be expressed in monopoly on some kinds of application of lawful violence (for example, in execution of judicial verdicts) or in monopoly on the authority sanction to apply force even if the verdict was carried out by the interested part, or in interdiction of some types of violence (for example, regarding a blood feud; see e. g., about the Aztec state … (pp. 120–121, emphasis in the original).

Similarly, Vliet, says:

But judicial procedures were established by the laws, the magistrate decided on the legal ground of the prosecution, and the courts were institutions of the state. … Prosecution were left to the individual citizen (ho boulomenos) also when the common interest of the polis was involved. The authorities as such did not prosecute ... That does not imply that that was less effective – rather the contrary, perhaps (p. 135).

Thus both Grinin and Vliet seem to take the fact that self-help, or self-regulation, was sanctioned by the courts and was carried out within the rule of the law, as an indication for the existence of the State which laid down the laws which qualified the use of private force. However, the existence of laws which sanction and qualify the use of self-help does not necessarily imply the existence of a State which legislates them: in (tribal) stateless societies self-help is not capricious but regulated by laws and customs. Further, it is also arguable to what extent the laws and customs which regulated the use of self-help in the Greek polis, were a product of the Athenian political institutions. As Hansen observes, in Athens ‘[t]he administration of justice concerning acts of violence and offences against property is … surprisingly archaic compared with the development of the (Athenian) political institutions...’ (Hansen 1976: 7, 113–121; Berent 1996: 58). This means that self-help was regulated by archaic customs rather then by laws laid down by the assembly or the polis. Further, as I have argued in detail elsewhere, the decentralized nature of Greek society and the absence of coercive apparatuses meant that the laws had to be identical with the customs of the community, or else that decisions had to be shared by a wide consensus. Thus to the extent that the laws which regulated the use of force could be seen as a product of polis's institutions, e.g. the assembly, the latter was not a ‘supreme power’ or ‘central authority’ which legislated or laid down norms, but rather reflected those already existent in the community (Berent 2000c).

According to Grinin and Vliet the role and boundaries of self-help were prescribed by ‘the State’, or in other words, self-help supplemented the State.

Thus Grinin says that

if the citizens could manage quite well the exploitation of slaves and keep them in obedience, why then the state should assume this function? … It follows that it was quite sufficient to have a state in Athens that sanctioned slavery and did not object to keeping the slaves subdued and managed by their masters … (p. 109).

Similarly Vliet explains why ‘the authorities’ did not involve in the execution of judicial sentences:

… Berent departs from a modern, or rather a modernistic view of what the law is and what the administration of justice should be. Among the ancient Greeks going to court was less a way of resolving conflicts between litigants than a way of continuing them. Thus, seeking justice was a series of connected conflicts.

Here the argument is as follows: the State did not intervene because there was no need to intervene. Thus in the case of the control of the slaves, the citizens managed it by themselves, or the State did not intervene in the execution of court judgment because the Greeks concept of courts and justice was different from the modern one and so, again, there was no need to interfere. But this is, indeed, to put the argument on its head. It is not the case that because Greek society happened to manage things by itself that there was not need for state intervention, but the other way around: Greek society was so organized (that is to manage things by itself) because it was stateless. Thus, for instance, the continuation of conflicts mentioned above by Vliet characterizes stateless communities in general because the regulation of conflicts by self-help is problematic. This is so because of the absence of a ‘neutral’ force which could enforce the resolution of the conflict. Consequently those who were trying to get by force what they considered was their due, even when granted by the courts, were likely to meet resistance. In the violent struggle which followed excesses occurred, also because there were private interests and passions involved, and the seeds of the next round of the conflict were sown (Hunter 1994: 140–143).

Conclusions – The Statelessness of the Greek Polis, why is it important?

What characterizes the controversy about the statelessness (or the statehood) of the Greek polis is the attempts of my critics to ‘prove’ that the Greek polis was a State. Indeed, this is one way to test and refute my theory (though, as noted above, the opposite is not true, that is the fact that the polis was not a state does not necessarily imply that it was stateless). Yet, the trouble with this kind of debate is that it tends to become a debate about definitions, and this, in turn, tends to become circular and futile. Norberto Bobbio has put forward the important question, ‘Did there exist a political society which could be called a “state” before the large territorial states with which the history of the modern State begins? and for which, therefore, any other specification is useless?’ (Bobbio 1989: 60). His position is that the answer to this question ‘depends entirely on the definition of the state’ and that ‘[t]he choice of definition depends on pragmatic criteria rather than truth’ (ibid.: 61). This would mean also that the choice of definition is, to a certain extent, arbitrary and that it is no doubt possible to find a definition of the State according to which the polis would be a State. Karl Popper offered a solution to this problem: we should start our debate with traditional theories and then amend them if necessary (Popper 1968: 120–135). The choice of traditional definitions of the State is pragmatic not only because it supplies us with a reasonable starting point but also because it could prove to be very fruitful. It is only within the traditional definitions of the State that the notion of the stateless polis challenges major academic disciplines and consequently could have a strong explanatory power. I tried to follow Popper's advice by choosing what had seemed to me to be the traditional theories of the State. I think that in the light of the latter the polis was not a State and I believe that both Grinin and Vliet seem also to suggest that the traditional theory of the early State is not easily reconciled with that of the Greek polis. Instead they offer the notion of the ‘early democratic state’ or that of the ‘citizens-state’. As I have noted above, though the latter might present an innovation in the anthropological study of the State, in the field of the classics they have been traditionally employed. This show us, pace Popper, that it is not very easy to determine what the traditional theories are and that sometimes we have in different fields different conflicting traditions. But to the extent that Vliet and Grinin indeed offer a divergence form the traditional concept of the (early) State, one might ask, why is it important? Why bother to prove that the polis was a State? This question could be leveled also against my notion of the stateless polis. If the traditional theory of the Greek polis considered the latter to be a State why is it important to conjecture that it was not a State and even stateless? Here I believe that one of the pragmatic criteria which should guide us in our choice of theories is what Karl Popper called the ‘explanatory power’ of the theory. Does it give a better explanation than the traditional theory? Does it explain certain things that the traditional theory could not explain? In my various papers I have tried to show how the notion of the stateless polis gives us a better understanding of the Greek polis and Greek political thought. It is interesting to note that both Grinin and Vliet, and even Hansen, do not seem to challenge my interpretations of the Greek experience, but only to point out that these do not ‘prove’ that the polis was not a state or that it was stateless, because in one way or another they characterize also state-societies. Yet, as I have said above, my interpretations of the Greek experience were not meant to ‘prove’ that the Greek polis was stateless, but they are only demonstrations or corroborations of its statelessness. The more there will be facts about the Greek polis which are conjectured by and reconciled with the notion of its statelessness the more the latter will be corroborated, or to use Popper's terminology, closer to the truth. Thus, though each and every one of my interpretations of the Greek experience (and all of them together) is objectionable as a ‘proof’ of the statelessness of the Greek polis, the important question is to what extent the overall picture which emerges as a result of the notion of the statelessness of the Greek polis is more clear than the one which previously prevailed. Thus from my point of view the theory of the statelessness of the Greek polis is not only ‘true’, but also fruitful. It gives us better and new insights of the Greek experience. It explains more facts about the latter, simplifies it and it makes it more coherent. In short, to use Popper's terminology, it has a greater explanatory power and is ‘a much better approximation of the truth’ than the traditional theory of the Greek State. Consequently I believe that the focus of the debate on the statelessness of the Greek polis should move from the attempts to ‘prove’ or disprove the statehood of the polis to attempts to examine to what extent the Greek experience corroborates the notion of the stateless polis? Or to what extent the latter gives us a better understanding of the Greek experience? Or, a better understanding of the State (ancient or modern or both)? And indeed, there are many issues to be discussed here (to some of them I have related in my previous papers): the relation between private and public or between morality and politics, Greek rule of law, Greek attitudes toward tyranny (or the use of force in politics), the nature of stasis and many more. When the focus of the debate would move in this direction, the notion of the stateless polis might prove to be a powerful tool for the understanding the ancient Greek world.


* Editors' note: The discussion on the nature of antique polity was started in volume 3, number 2 (September2004) of our Journal by Leonid E. Grinin in his article titled ‘Democracy and Early State’. Then the complicated issue, namely whether the Greek polis and the Roman Republic were early states or non-state complex societies of a specific type, was considered by Henri J. M. Claessen in his contribution ‘Early State Intricacies’ and Edward van der Vliet in his ‘Polis. The Problem of Statehood’ (both published in Volume 4, number 2 [September 2005]).

1 I have dealt with this issue extensively in Berent, ‘Anthropology and the Classics’.


Barker, E.

1960. Greek Political Theory. Plato and His Predecessors. London: Methuen's University Paperbacks.

Berent, M.

1994. The Stateless Polis: Towards a Re-Evaluation of the Classical Greek Political Community. Dissertation. Cambridge.

1996. Hobbes and the ‘Greek Tongues’. History of Political Thought XVII (1): 36–59.

1998. Stasis, or the Greek Invention of Politics. History of Political Thought XIX (3): 331–362.

2000a. Anthropology and the Classics: War, Violence and the Stateless Polis. Classical Quarterly 50 (1): 257–289.

2000b. Greece (11th–4th centuries B. C.). In Bondarenko, D. M., and Korotayev, A. V. (eds.), Civilizational Models of Politogenesis (pp. 228–251). Moscow: IAf RAN.

2000c. Sovereignty: Ancient and Modern. Polis 17 (1–2): 2–34.

2000d. The Stateless Polis: the Early State and the Ancient Greek Community. In Kradin N. N. et al. (eds.), Alternatives of Social Evolution (pp. 225–241). Vladivostok: FEB RAS.

2004a. Greece: The Stateless Polis (11th – 4th Centuries B. C.). In Grinin L. E. et al. (eds.), The Early State, Its Alternatives and Analogues (pp. 364–387). Volgograd: Uchitel.

2004b. In Search of the Greek State: A Rejoinder to M. H. Hansen's. Polis 21.

Bobbio, N.

1989. Democracy and Dictatorship. Cambridge: Polity Press.

Crone, P.

1986. The Tribe and the State. In Hall, J. A. (ed.), States in History (pp. 48–77). Oxford: Blackwell.

d'Entrèves, A. P.

1967. The Notion of the State. An Introduction to Political Theory. Oxford: Clarendon.

Ehrenberg, V.

1969. The Greek State. 2nd ed. London.

Finley, M. I.

1983. Politics in the Ancient World. Cambridge.

1985a. The Ancient Economy. 2nd ed. London.

1985b. Democracy Ancient and Modern. 2nd ed. London: Hogarth Press.

Fortes, M., and Evans-Pritchard, E. E.

1940. Introduction. In Fortes, M., and Evans-Pritchard, E. E. (eds.), African Political Systems. Oxford.

Gellner, E.

1983. Nations and Nationalism. Oxford: Blackwell.

Grinin, L. E.

2004. Democracy and Early State. Social Evolution and History 3 (2): 93–147.

Hansen, M. H.

1976. Apagoge, Endeixis and Ephegesis against Kakourgoi, Atimoi and Pheugontes. Odense.

1991. The Athenian Democracy in the age of Demosthenes.

1998. Polis and City-State. An Ancient Concept and its Modern Equivalent. CPCActs (pp. 89–91). Copenhagen.

2002. Was the Polis a State or a Stateless Society? In Nielsen, Th. H. (ed.), Even More Studies in the Ancient Greek Polis (pp. 17–47). Papers from the Copenhagen Polis Centre 6. Stuttgart.

Hunter, V. J.

1994. Policing Athens. Social Control in the Attic Lawsuits, 420320 B.C. Princeton, New Jersey: Princeton University Press.

Khazanov, A. M.

1978. Some Theoretical Problems of the Study of the Early State. In Claessen, H. J. M., and Skalnik, P. (eds.), The Early State (pp. 77–92). The Hague: Mouton.

Ober, J.

1989. The Nature of Athenian Democracy: a Review of ‘The Athenian Assembly in the Age of Demosthenes’ by M. H. Hansen. Classical Philology 84: 322–334.

Popper, K.

1968. Towards a Rational Theory of Tradition. In Conjectures and Refutations: The Growth of Scientific Knowledge (pp. 120–135). New York.

Runciman, W. G.

1990. Doomed to Extinction: the Polis as an Evolutionary Dead End. In Murray, O., and Price. S. (eds), The Greek City from Homer to Alexander (pp. 347–367). Oxford.

Sahlins, M.

1968. Tribesmen. Englewood Cliffs, New Jersey.

Skinner, Q.

1989. The State. In Ball, T., Farr, J., and Hanson, R. L. (eds.), Political Innovation and Conceptual Change (pp. 90–131). Cambridge: Cambridge University Press.

van der Vliet, E. Ch. L.

2005. Polis. The Problem of Statehood. Social Evolution & History 4 (2): 120–150.